| Project/Area Number |
17K20024
|
|---|
| Research Category |
Grant-in-Aid for Challenging Research (Exploratory)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Applied informatics and related fields
|
|---|
| Research Institution | The University of Electro-Communications |
|---|
Principal Investigator |
Uto Masaki 電気通信大学, 大学院情報理工学研究科, 助教 (10732571)
|
|---|
| Project Period (FY) |
2017-06-30 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥6,500,000 (Direct Cost: ¥5,000,000、Indirect Cost: ¥1,500,000)
Fiscal Year 2019: ¥2,340,000 (Direct Cost: ¥1,800,000、Indirect Cost: ¥540,000)
Fiscal Year 2018: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
Fiscal Year 2017: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
|
|---|
| Keywords | 自動採点 / 深層学習 / トピックモデル / テスト理論 / 論理マイニング / 言語処理 / 小論文試験 / 論理性 / ディープラーニング / 議論マイング / 論理構造 / ベイジアンネットワーク / 自然言語処理 / 論証マイニング / 統計的機械学習 / ベイズ統計 / 小論文自動採点 |
|---|
| Outline of Final Research Achievements |
With an increasing need for large scale essay-writing tests, automated essay scoring (AES), which utilizes natural language processing and machine learning techniques to grade essays automatically, has been attracted wide attention. This study aimed to develop a new AES method focusing on “argument structure” by combining argument mining techniques, which is a state-of-the-art NLP technique. However, this approach did not improve the performance of AES sufficiently. On the other hand, during this research process, we found that conventional AES methods share the same bias factor, which may cause considerable performance degradation. Thus, this study developed a new AES method that can deal with the bias problem. The proposed method achieved a significant improvement in the AES performance. The results have been accepted in an academic journal and a top international conference.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究では，自動採点モデルへの組み込みが難しいとされてきた「論理性」を明示的に考慮することで自動採点の精度向上を目指した．しかし，近年の自動採点技術の高度化の影響もあり，この導入が必ずしも十分な性能改善に寄与しないことが明らかになった．他方で，本研究の過程で発見した「自動採点技術に共通するバイアス」の問題は，本質的でありながら，これまでは無視・軽視されてきた点であり，学術上も実用上も重要な指摘といえる．小論文自動採点技術は，実用化が強く望まれるにも関わらずその高精度化が困難な技術の一つであり，本研究における発見と進展は，挑戦的研究としての学術的にも社会的にも意義のあるものといえる．
|
